Butterfly Network Receives First FDA Clearance for AI-Powered Gestational Age Ultrasound Tool
核心洞察
Butterfly Network secured FDA clearance for the first AI-powered ultrasound tool to estimate gestational age in the U.S., delivering results in under two minutes without requiring image capture or biometric measurements.
The tool was trained on over 21 million ultrasound images from diverse populations and provides consistent results for pregnancies between 16-37 weeks using a "blind-sweep" method.
The technology addresses critical healthcare gaps in emergency settings, rural U.S. communities, and low-resource international markets where maternal mortality remains high.
Butterfly Network announced on Monday that it received U.S. Food and Drug Administration clearance for an artificial intelligence-powered ultrasound tool that estimates gestational age, marking the first FDA approval for such technology in the United States. The clearance positions the company to expand access to maternal care in underserved areas where traditional ultrasound services are limited.
Revolutionary AI Technology
The AI tool delivers gestational age estimates in under two minutes without requiring users to capture or interpret images or perform fetal biometric measurements. Unlike traditional ultrasound machines that rely on expensive piezoelectric crystals, Butterfly's device uses a single silicon chip for whole-body imaging, making it more accessible and cost-effective.
The model was trained on more than 21 million ultrasound images from diverse patient populations and clinical settings. According to Butterfly, the tool provides "consistent results in patients between 16-37 weeks that are equivalent to sonographer-performed, biometry-based gestational age assessments."
The technology uses an AI-based "blind-sweep" method designed to minimize user variability and provide swift results, making it particularly valuable for clinicians without specialized ultrasound training.
Addressing Critical Healthcare Gaps
Steve Cashman, Butterfly's chief business officer, explained that "the GA tool is built to address three critical use cases: enabling faster decisions in emergency settings when gestational age is unknown; expanding access in countries where maternal and neonatal mortality remains high; and improving prenatal ultrasound access in rural US communities, where patients may otherwise travel hours for."
The tool addresses significant gaps in prenatal care access. Research indicates that almost half of rural U.S. areas lack hospital obstetric services, while 92% of maternal and neonatal deaths globally occur in low- and middle-income countries, many of which could be prevented with access to basic imaging and earlier intervention.
Global Deployment and Expansion Plans
Butterfly has already deployed the gestational age tool in Malawi and Uganda through an initiative made possible in part by a grant from the Gates Foundation (搜索). The FDA clearance will enable the company to expedite rollouts across additional regions in sub-Saharan Africa and expand use in the United States.
The tool is integrated into Butterfly's app and is intended to help clinicians make faster decisions in emergency departments, rural clinics, and other low-resource settings. Cashman emphasized that the technology is "about helping clinicians act faster, earlier, and with greater confidence, at critical moments and in any setting."
Regulatory Context
The clearance aligns with the FDA's encouragement of artificial intelligence in medical devices. The FDA states on its website that AI and machine learning have the potential to improve healthcare by generating new insights from data produced during care delivery.
Gestational age serves as the standard measurement of pregnancy progression and helps clinicians time key prenatal screenings, detect complications sooner, guide interventions, and improve outcomes for mothers and babies. For communities facing socioeconomic barriers and lack of access to basic obstetric imaging protocols, establishing this metric can be particularly challenging.